It's difficult to say what the best approach is for your situation here, as I cannot tell what your diffusers look like, what they are made of, or how you expect them to work. In a general sense however, it is true (as stated in the previous post you mention) that SketchUp and OpenStudio (and EnergyPlus and even Radiance to an extent) do not support true curves (e.g. cylinders, discs, rings). These shapes are approximated via segmentation, and you should use as few segments as needed to suggest the behavior of the diffuser design.

As far as modeling light-blocking or -redirecting "bits" in the interior space, from a Radiance simulation perspective, I would recommend using interior partition surfaces. As these are only translated to Radiance, you can use as many surfaces to represent your design as needed.

You also mention the need to translate transmittance; presumably your diffuser has transparent or translucent elements to it? Currently the Radiance translator assumes shading and interior partition surfaces are opaque, and subsurfaces are not handled. We can certainly add support for this, it's merely a question of how best to approach it. I'm hoping others on the team can weigh in here with thoughts on how best to handle this.

In the meantime, could you flesh out your question, perhaps with an image or sketch of the diffuser design(s)? That would help guide us on the path to the best solution from the model/translator side of things.

I discarded the option of modeling a diffuser with curvatures. In the following figures are shown the diffusers i was trying to model:

https://www.dropbox.com/s/ksea50ko8mv...

In real life the hole in the ceiling of the room is circular, but for simulations I decided to switch it for a square hole with same area, as shown in the following figure:

https://www.dropbox.com/s/ig427yt96qm...

Finally I modeled a simple square diffuser with a planar surface constructed with "Partition Interior Surface Group" and I could assign parameters of absorptance and reflectance, as shown in the following figure:

https://www.dropbox.com/s/d7n3aspl07u...

Thanks for the suggestions and help.

I have a similar question related to the type of material. I am trying to find methodologies to model the material similar to interior wait curtains or diffusing panels like:

How can we define the level of diffusing "intensity"? Or is it only possible to work with transmittance and reflectance? My model is predicting high values of illuminance and I would like to diffuse the solar rays.

I was checking "glaze" script but, from my understanding, it only includes the inner ray reflectances within a glazing system.

May I ask you what type of material did you use?

Thank you in advance